New Insights on Insurance-Linked Securities, Risk-Adjusted Performance Measurement and Behavioral Insurance

All three papers that are part of this research proposal strongly rely on insights from my own extant work that was supported by a one-year postdoctoral grant from the fundamental research fund of the University of St. Gallen (GFF-HSG) between mid-2012 and mid-2013. The first paper focuses on the market for cat bonds and thus directly follows one of the topics from the previous proposal. More specifically, I would now like to examine whether cat bonds (still) offer an adequate risk premium for investors. Since historical return time series for the asset class are still way too short to include the impact of mega-events with reoccurrence periods of more than 100 years, this question can only be adequately answered by benchmarking a modeled return distribution with historical data from the stock and bond markets. For this purpose I will draw on RMS Miu, a sophisticated software package that is used for catastrophe risk modeling in the (re)insurance industry. Based on my results, it will be possible to close a critical gap in our understanding of the insurance-linked securities asset class. The second paper, in contrast, draws on the modeling expertise that I gained during the work on financial regulation that was part of my last proposal. More specifically, my coauthors and I developed a solvency model for insurance companies which will now form the basis for a whole new performance measure in the industry. This measure will be innovative in that it takes into account the market risks both on the asset and the liability side of an insurer's balance sheet. In doing so, it attributes a higher performance to companies that achieve their targeted return with a less volatile equity capital. Finally, in the third paper I want to estimate a comprehensive Bayesian structural equation model to explain the willingness to pay for term life insurance. For this purpose, individual-level part-worth utility curves that have been measured by means of choice-based conjoint analysis will be combined with demographic, socioeconomic, and latent factors. Since all three papers address pressing questions that can be expected to exhibit a notable impact on the insurance industry and the capital markets in the near future, I believe that this fundamental Research project is both relevant and timely.